Market and Technology Trends
Automotive ADAS 2026
ADAS growth is shifting from sensor expansion to compute, software, and integration, while China emerges as a separate ecosystem.
YINTR26562Key Features
ADAS revenue rises from $20.4B in 2021 to $66.3B in 2031. Growth is shifting from sensor proliferation to compute, software, and scalable ADAS platforms, with China as the main growth engine.
ADAS supply chains are shifting from module-based sourcing to control of compute, software, and system integration. OEMs are working more directly with SoC vendors, Tier-1s are moving up the stack, and China is building a distinct local ecosystem.
ADAS technology is converging toward centralized, AI-driven architectures, with smarter cameras, 4D/imaging radar, long-range LiDAR, stronger compute, faster networks, and rising memory needs enabling richer functions.
Report objectives
Deliver an overview of the automotive ADAS market, with forecasts by application:
- ADAS market by Tier-1 and by region.
- Market forecasts in units and revenue.
- Split by sensing, compute, software, and system integration.
Explain how ADAS architectures are evolving across sensors, compute, and software:
- Trends in camera, radar, LiDAR.
- Evolution from distributed ECUs to centralized and domain-based architectures.
Provide an in-depth understanding of the competitive ecosystem:
- Positioning of OEMs, Tier 1s and how their role is evolving.
- Focus on Chinese ecosystem
Offer key technical insights and analyses into future technology trends and challenges:
- Key technology choices.
- Technology dynamics and roadmaps.
- Chiplet for automotive.
From sensor proliferation to platform value
The ADAS market is entering a new expansion phase, which should last through 2031, with growth no longer driven by sensor proliferation alone but increasingly by architectures with more centralized computing, and software content.
- Total ADAS (sensors and ECUs) revenue rises from $20.4B in 2021 to $66.3B in 2031, with the fastest growth coming from ECUs and software. ECU revenue outpaces sensors revenue, as domain and centralized architectures scale up.
- Sensor growth remains strong, but the mix shifts toward higher-value technologies. Cameras remain the volume backbone; radar transitions from legacy products to 4D and imaging solutions; and LiDAR shows the fastest growth, though from a small base, driven mainly by long-range ADAS deployments.
- Within sensing, value creation increasingly comes from smarter and more differentiated configurations. In cameras, front smart cameras and 360° surround remain the largest revenue pools, while conventional rear-view modules decline structurally. In radar, 4D becomes the dominant segment, and in LiDAR, long-range sensors continue to anchor volume through 2031.
- China emerges as the clearest growth engine and reshapes the competitive landscape. Rapid L2+/NOA adoption and a strong domestic ADAS supply chain support faster market growth and accelerate the rise of Chinese Tier-1s, with Momenta and Huawei in particular gaining share alongside other domestic suppliers.
Overall, future ADAS growth will increasingly depend on the ability to orchestrate sensors, compute, and software into scalable ADAS platforms.
Compute is redrawing the ADAS supply chain
The ADAS supply chain is entering a new phase, with value no longer centered on standalone sensing modules only but increasingly on computing platforms, software integration, and control of system architecture. This is reshaping relationships among OEMs, Tier-1 suppliers, and semiconductor companies while accelerating the divergence between Chinese and Western supply chains.
- Leadership remains highly layer-specific across the ADAS stack. Competitive positions still differ materially across processors, radar, camera, and LiDAR, reflecting distinct technology cycles, cost structures, and customer concentration by product category. No supplier dominates the full stack.
- Compute is changing both sourcing logic devices and the role of Tier-1s. As OEMs build closer ties with SoC vendors, the traditional linear supply model is weakening. Tier-1s are increasingly pushed beyond hardware delivery toward compute integration, software enablement, validation, safety, and industrialization.
- China is developing a more localized and vertically coordinated supply-chain model. Faster feature rollout, stronger domestic ecosystems, and integrated pilot packages are accelerating the localization of compute, sensors, and software, with LiDAR leading and camera and radar increasingly following.
- Global ADAS sourcing is becoming more regionalized and strategic. Geopolitics, data security, trust requirements, and ecosystem differences are driving separate supply-chain configurations for Chinese and Western markets, while in-house chip strategies are expanding selectively among OEMs seeking greater control over compute roadmaps.
Overall, ADAS supply chains are being restructured around control of compute, software, and architecture, with regional divergence becoming a defining feature of the market.
Sensor roadmaps diverge, architectures converge
The next phase of automotive sensing will depend less on sensor improvements alone and more on how each sensor fits into a more centralized ADAS architecture. As perception becomes more compute-intensive and software-driven, sensor value will increasingly come from integration, scalability, and system-level contribution.
- Camera innovation is broadening beyond core CIS metrics. Resolution, HDR, sensitivity, and dynamic range remain important, but the next wave of differentiation also comes from improved connectivity, hybrid optics, and active imaging approaches such as gated imaging.
- Radar is moving from a detection sensor to a richer perception layer. Radar megatrends are being enabled by advances in RF silicon, higher integration, and increasing centralization of computing, allowing better angular resolution, more advanced processing, and a stronger role in multi-sensor perception architectures.
- LiDAR remains on a dual-technology roadmap. ToF dominates the market today because it offers the most mature balance of performance, cost, and industrial readiness, while FMCW is positioned as the next technical step, first in niche and premium applications before broader automotive adoption.
- Global ADAS sensor roadmaps remain heterogeneous. Sensor architectures continue to differ by OEM strategy, autonomy level, vehicle segment, and region, but the common direction is clear: sensors are becoming more tightly integrated with centralized computing, software-defined perception, and system-level optimization.
Overall, the next phase of sensor competition will depend less on isolated device specifications than on each modality’s ability to deliver scalable performance within increasingly centralized, compute-intensive ADAS architectures.
Glossary
Global semiconductor activity
Objectives of this report
Scope of this report
Methodology & definitions
Companies cited
3-page summary
Executive summary
Context
Market forecasts
- Tier-1 ADAS revenue and market share
- Image sensor and camera module in Munits & $M
- Radar module in Munits & $M
- LiDAR module in Munits & $M
- Computing ADAS in Munits & $M
Market trends
- AEB discussion
- L2+ is mainstream now
- L3 remains in the light of sight for OEMs
- Sensor and computing needs per autonomy level
- Physical AI: why it matters for automotive ADAS
Market shares and supply chain
- ADAS Ecosystem
- Market shares (imaging / radar / LiDAR / Computing)
- How Tier1s are building a one-stop ADAS offer
- How do OEMs make their own chips?
- Make vs Buy Automotive chip – Economic perspective
- Chinese Landscape
Technology trends
- From smart sensors to centralized perception
- The end-to-end (E2E) perception pipeline: where sensors meet compute
- SoC technology nodes (5nm and below / chiplets)
- From E2E stack requirements to E/E architecture enablers
- Memory becomes the next ADAS bottleneck
- Global ADAS roadmap
Technology, process, and cost report
- Teardown sample (Camera, LiDAR, radar, domain controller)
Outlook
About Yole Group
Aeye, Aeva, Aisin, Alps Alpine, Ambarella, AMD, Analog Photonics, Aptiv, Arbe, Arcfox, Aumovio, BAIC Group, Black Sesame, BMW Group, Bosch, Brose, BYD Auto, Changan Automobile Group, Chery Automobile, Deepal, Denso, Desay SV, Dongfeng Motor Corp., Echodyne, China FAW Group Corp., Ficosa, Ford Group, Forvia, Freetech, GM Group, Great Wall Motor Company Ltd., Guangzhou Automobile Group, Hailo, Hasco, Hesai, HikVision, HiSilicon, Hitachi Astemo, HL Mando, Honda, Horizon Robotics, Huawei, Hyundai Kia Automotive Group, Infineon, InnoSenT, Kyocera, Li Auto, Link & Co, LG Innotek, Lucid, Lynred, Magna, Mahindra & Mahindra, Mazda, MediaTek, Mercedes-Benz, MicroVision, Mobileye, Nanoradar, Nidec, Nio, Nvidia, NXP, Omnivision, ON Semiconductor, Panasonic, Polestar, Qualcomm, Renault-Nissan-Mitsubishi Alliance, Renesas, Rivian, RoboSense, SAIC, Samsung, Scantinel Photonics, SemiDrive, Seyond, SiEngine, SiLC, Sony, Steerlight, Stellantis, STMicroelectronics, Subaru, Suzuki, Tata Group, Tesla, Texas Instrument, Toshiba, Toyota Group, TSMC, Uhnder, Valeo, Volvo, Voyant Photonics, VW Group, Waymo, Weifu, WHST, Xiaomi, Xpeng, Zeekr, ZF, Zhejiang Geely Holding Group, ZYT and more.